- Email: [email protected]
Human platelet phenolsulphotransferase: Separate control of the two forms and activity range in depressive illness
333
Clinicu Chimicu Actu, 117 (1981) 333-344 Elsevier/North-Holland Biomedical Press
CCA
1978
Human platelet phenolsulphotransferase: separate control of the two forms and activity range in depressive illness Susan
M. Bonham
Carter
a, Vivette
P.K. Gillman
Glover
b and P.K.
a, M. Sandler
Bridges
a,*,
b
u Bernhurd Buwn Memoriul Reseurch Luhorutories und Institute of Obstetrics and Gynuecologv, Queen Charlotte’s Hospitul, Goldhuwk Roud, London Wci OXG (UK) und’ The Geoffrey Knight Psychosurgicul Unit, Brook Generul Hospitul, Shooters Hill Roud, London SE18 4L W (UK) (Received
June 16th. 198 I)
Summary Human phenolsulphotransferase exists in two forms, one specific for dopamine and tyramine, termed “M” and one for phenol, termed “I”‘. In this study we have shown that these two forms are under separate control by correlating their activities in different individuals using different substrates. There was a highly significant correlation between the activities with dopamine, p-tyramine and 4-hydroxy-3methoxyphenylglycol, but no significant correlation between the activities with any of these three substrates and that with phenol. Neither age nor sex had any effect on platelet phenolsulphotransferase “M” or “P” activities. Nor was there any significant correlation between platelet monoamine oxidase activity and phenolsulphotransferase “M” or “I”’ activities. Human platelet phenolsulphotransferase “M” was found to be unstable at temperatures above 35°C and it lost substantial activity when stored deep frozen in isotonic saline. However it was stable for up to four months when stored in isotonic sucrose or 10 mmol/l phosphate buffer (pH 7.4). Phenolsulphotransferase “M” and “I”’ activities were measured in platelets from depressed patients of a diagnostic type characterized by low output of tyramine-osulphate after oral tyramine loading but their enzyme activities were not different from those in two control groups.
Introduction Depressed patients tend to excrete less conjugated tyramine (tyramine-0-sulphate) after an oral load than normal subjects [ 1,2] and this defect is trait- rather than * To whom correspondence 0009-8981/81/0000-0000/$02.75
should be addressed. 0 1981 Elsevier/North-Holland
Biomedical
Press
334
state-dependent [2,3]. One possible cause is a decrease in activity of phenolsulphotransferase (PST) (EC 2.8.2.1) the enzyme which catalyzes the transfer of sulphate from 3’-phosphoadenosine-S-phosphosulphate (PAPS) to a phenolic acceptor [4]. PST has recently been identified in human platelets [5-71, thus opening the way to a wide variety of clinical studies. In this present work, we have measured PST activity in severely depressed patients, patient controls and staff controls. Because little work has so far been done with human PST, it was first necessary to establish the best conditions for platelet preparation and storage prior to enzyme assay. Sulphate conjugation catalyzed by PST is an important metabolic pathway, not only for phenolic monoamines and some of their metabolites but also for phenol and some phenolic drugs [8]. Rein et al. [9] have recently found evidence for the existence of two forms of human PST, one specific for phenol (named PST “P”) and one for tyramine and dopamine (PST “M”). They differ in tissue distribution and inhibitor sensitivity. In order to determine whether these two forms are controlled together or independently, we have determined the degree of correlation between platelet PST activities in individual subjects using tyramine, dopamine, 4 - hydroxy - 3 methoxyphenylglycol (HMPG) and phenol as substrates. Materials [35S]PAPS (1.9-4.2 Ci/mmol; New England Nuclear Corporation, Boston, MA, USA) was shipped in dry ice and stored in 50% ethanol at -20°C. Non-radioactive PAPS (PL Biochemicals, Inc., Milwaukee, WI, USA) was shipped and stored similarly. p-Tyramine HCl, dopamine HCl, 4-hydroxy-3-methoxyphenylglycol piperazine salt (HMPG), dithiothreitol (DTT) and bovine serum albumin (BSA) were purchased from Sigma Chemical Company, Poole, Dorset, UK. Phenol was obtained from BDH Chemicals Ltd., Poole, Dorset, UK and Instagel from Packard Instrument Co., Wembley, UK. Methods Sample
collection and platelet preparation
For stability studies: venous blood samples (10 ml) were collected from normal volunteers into plastic Universal containers with 0.5 ml 5% ethylenediaminetetraacetic acid (EDTA) and mixed gently. They were centrifuged at 327 X g for 6 min at room temperature; platelet-rich plasma was carefully removed by aspiration and samples from groups of three or four subjects were pooled before being divided equally into three and spun again at 2500 X g for 20 min at room temperature. The three platelet buttons so obtained were treated differently; one was resuspended in isotonic saline (0.15 mol/l; 1 ml/10 ml blood) and two in isotonic sucrose (0.3 mol/l; 1 ml/10 ml blood) before transfer to another Universal container, care being taken to leave any red blood cells behind. The suspensions were recentrifuged at 2500 X g for 20 min. The platelet button that had been washed with saline was
335
drained and resuspended in isotonic saline (1 ml/ 10 ml blood); one of the preparations washed with isotonic sucrose was resuspended in isotonic sucrose (1 ml/10 ml blood) and the other in phosphate buffer (10 mmol/l, pH 7.4; 1 ml/10 ml blood). These suspensions were then divided into aliquots as required before storage at - 20°C. For clinical studies: blood was obtained from 18 patients suffering from therapyresistant depressive illness, of the same diagnostic category in which we had already demonstrated a decreased tyramine-conjugating ability [2], from 18 non-psychiatric patient control subjects and 14 staff control subjects. Blood samples from 30 of the subjects (10 depressed patients, 10 patient controls and 10 staff controls) were collected on one day and processed together, and from the remaining 20 subjects (8 depressed patients, 8 patient controls and 4 staff controls) on another day, again being processed together. Venous blood (10 ml from each subject) was collected into plastic Universal containers with 0.5 ml 5% EDTA and mixed gently. It was centrifuged at 327 X g for 6 min, the platelet-rich plasma being carefully transferred to a new Universal container and the platelets spun down at 2500 X g for 20 min. Platelet buttons were drained, resuspended in isotonic sucrose (1 ml/10 blood) and transferred to new Universal containers, care being taken to leave any red blood cells behind (washing step). The platelets were spun down again at 2500 X g for 20 min. After the supernatant had been removed, the platelets were finally resuspended in phosphate buffer (10 mmol/l, pH 7.4; 1 ml/ 10 ml blood) and stored at - 20°C until assay. PST
assay
PST activity was measured by the radioenzymatic method of Foldes and Meek [lo], using radioactive [35S]PAPS as sulphate donor. For stability studies, the incubation mixture consisted of 100 ~1 potassium phosphate buffer (10 mmol/l, pH 7.4) 10 ~1 pooled platelet suspension and 20 ~1 tyramine solution (1 mmol/l, giving a final concentration of 133 pmol/l in the incubation mixture) or 20 ~1 water for the blanks. All samples were assayed in duplicate with a blank for each pooled platelet sample. The reaction was initiated by adding 20~1 of a 4.5 pmol/l solution of [35S]PAPS and non-radioactive PAPS in water (final concentration in incubation mixture 0.6 pmol/l) to successive tubes at 10 s intervals. After incubating for 10 min at 37°C the reaction was stopped at 10 s intervals by adding 200 ~1 barium acetate (0.1 mol/l) and transferring both from the water to ice. Unreacted PAPS was removed by two successive precipitations with barium hydroxide (200~1, 0.1 mol/l) and zinc sulphate (200 ~1, 0.1 mol/l), the second being carried out on the supernatant obtained from spinning down the first. After centrifugation, the final supernatant was removed to a counting vial insert; Instagel (2.5 ml) was added and mixed to form a gel and the radioactivity measured in a Packard Tricarb liquid scintillation counter. Results were expressed as nmol product (tyramine-O-sulphate) formed/mg protein/l0 min incubation. For the clinical study, the assay procedure was the same but it was also performed on the platelet samples using acceptor substrates other than tyramine in the following final concentrations: phenol at 30 pmol/l, dopamine at 30 pmol/l and
336
HMPG at 800 pmol/l; 20~1 platelet suspension was used in the assay with phenol because of lower specific activity. Blanks containing 204 of water instead of acceptor substrate were assayed for each platelet sample. This was essential as blank values varied from individual to individual, although they were no more than 20% of the total using tyramine as substrate. All assays were carried out in duplicate, with the operator blind to the clinical data. Monoamine oxidase (MAO) assay MAO activity towards tyramine was measured in the platelets obtained from the clinical study, using a standard radioenzymatic method with [‘4C]tyramine as substrate (final concentration 150 pmol/l). Protein determination Protein concentrations of platelet suspensions were measured by the method of Lowry et al [l l] with BSA as standard. Results
Linearity with time and tissue concentration Platelet PST activity towards tyramine increased linearly with time up to 10 min incubation at 37°C. A 10 min incubation period was therefore used in all assays. Activity also increased linearly with increasing tissue concentration up to a value of
-60
-50
-40
-30
-20
-10
IO
20
30
Fig. 1. Double reciprocal plots of human platelet PST activity versus tyramine concentration at constant concentrations of PAPS: 0 0.06 pmol/l; 0 0.12 pmol/l; n 0.24 gmol/l; A 0.6 pmol/l and H 1.2 amol/l. K, for tyramine obtained from this experiment= 15 pmol/l; a second experiment gave similar results. Lines were reconstructed and the K, calculated by the direct linear plot method [121.
337
60 pg protein in the incubation mixture. The actual amount of protein used in assays ranged from IO-60 pg. K,
determinations
Human platelet PST activity was measured using different concentrations of tyramine and PAPS. The results are shown in Figs. 1 and 2. Apparent K, values were calculated by the direct linear plot method [ 121.The apparent K, for tyramine was 17 pmol/l (mean of two experiments), a value somewhat lower than that obtained by others, using Lineweaver-Burk plots [5,7]. The apparent K, for PAPS was 0.14 pmol/l (mean of two experiments), very similar to that obtained by Anderson and Weinshilboum [6] who also used the direct linear plot. Effect of BSA, of DTT and of washing the platelet preparation
Addition of BSA (final concentration 0.042%) and DTT (final concentration 8 mmol/l) to the incubation mixture had no significant effect on PST activity towards tyramine of pooled human platelets suspended in isotonic saline, isotonic sucrose or 10 mmol/l phosphate buffer. However, as will be seen in the section on stability with storage, DTT did sometimes partially restore PST activity which had been lost when the suspension was stored deep frozen in isotonic saline. Washing the platelet button, as described in the “Methods” section, was found to be essential to remove plasma proteins.
&
(gmol/l-‘)
Fig. 2. Double reciprocal plots of human platelet PST activity versus PAPS concentration at constant concentrations of tyramine: 0 33 pmol/l; 0 67 pmol/l; n 167 pmol/l. K, for PAPS obtained in this experiment=O. I5 pmol/l; a second experiment gave similar results. Lines were reconstructed and the K, calculated by the direct linear plot method [ 121.
Fig. 3. Thermostability of human platelet PST using tyramine as substrate. Platelets saline, n ; isotonic sucrose, A ; and phosphate buffer (10 mmol/l, pH 7.4), 0.
suspended
in isotonic
Freezing and thawing Different aliquots of the same pooled platelet sample, suspended in isotonic saline, isotonic sucrose or 10 mmol/l phosphate buffer, were frozen and thawed up to five times before assay for PST activity towards tyramine. Samples suspended in saline lost up to 50% of activity after such treatment and it was not restored by adding DTT (final concentration 8 mmol/l) to the incubation mixture. Platelets suspended in sucrose lost up to 25% activity after this repeated manoeuvre and, again, DTT did not restore lost activity. Platelets stored in 10 mmol/l phosphate buffer, however, lost no appreciable PST activity after successive freezing and thawing. Thermostability Aliquots of pooled platelet samples suspended in isotonic saline, isotonic sucrose and 10 mmol/l phosphate buffer were exposed to different temperatures in a water bath for 20 min prior to PST assay using tyramine as substrate. The results are shown in Fig. 3. The enzyme loses most of its activity at 40°C and a substantial amount even at 35°C in saline, sucrose or buffer; again, the addition of DTT to the incubation mixture failed to restore lost activity. Thus, it is obvious that great care must be taken when preparing platelets for PST studies, not allowing them to overheat, for instance, during centrifugation.
339 PST activity I-
60
80
100
(days)
Fig. 4. Stability of human platelet PST activity with storage time, using tyramine as substrate. Platelets suspended in isotonic saline, m; isotonic sucrose, A; and phosphate buffer (IO mmol/l, pH 7.4) l .
Stability with storage time Successive aliquots of pooled platelet samples stored deep frozen (-20*C) in isotonic saline, isotonic sucrose or 10 mmol/l phosphate buffer were assayed for PST activity using tyramine as substrate over a period of 116 days. No aliquot was thawed more than once prior to assay. The results are shown in ‘Fig. 4. It will be seen that, whereas samples stored in sucrose and phosphate buffer were stable over that time period, those stored in saline lost more than 85% of their activity after 56 days. The addition of DTT (final concentration 8 mmol/l) to the incubation mixture had no effect on PST activity of platelets stored in sucrose or buffer and only partially restored the lost activity of platelets stored in saline. PST activity in depressive illness Once the stability of human platelet PST in the three different platelet suspension media had been established, blood samples were collected from severely depressed patients and controls and platelet suspensions prepared and stored as described in the “Methods” section. PST activity was assayed in all 50 samples using tyramine TABLE
I
HUMAN = SEM)
PLATELET
PST ACTIVITIES
Substrate
(nmol product
formed/mg
protein/
IO min incubation:
Staff controls
Patient controls
Depressed patients
Tyramine
(133tr,mol/l)
0.39*0.03 (n=14)
0.39~0.02 (n=l8)
0.38*0.02 (n=ltl)
Dopamine
(30 pmol/l)
0.28t0.03 (?I= IO)
0.28r0.01 (n= 10)
0.23 20.02 (n=9f
0.33 eo.03 (n= IO)
0.32*0.02 (n=lO)
0.32 ItO. (I?= IO)
0.07 k 0.01 (n= 14)
0.07~0.01 (n=l@
0.07 2 0.01 (n== 18)
HMPG
(800 amol/l)
Phenol (30 amol/I)
* pCO.05
compared
to patient
controls
(Student
r-test).
*
mean
340
TABLE
II
CORRELATION INDIVIDUALS
COEFFICIENTS FOR HUMAN PLATELET WITH DIFFERENT SUBSTRATES
PST ACTIVITIES
IN DIFFERENT
Substrates
Tyramine
Dopamine
HMPG
Phenol
Tyramine
_
0.78 * (n=29)
0.88 *
0.27
(n=30)
(n=SO)
0.78 *
0.82 *
0.19
(n=29)
(n=29)
(n=29)
_
0.24 (n=30)
Dopamine
HMPG
0.82 * (n=29)
0.88 * (n=30)
* p
III
HUMAN VALUES
PLATELET PST: ACTIVITY RANGES FOR MALES AND FEMALES
Substrate (final concentration)
PST activity (nmol product Activity
range
formed/mg
FOR
DIFFERENT
protein/
SUBSTRATES
IO min)
Mean -ISEM males
females
Tyramine (133 pmol/l)
0.20-0.63 (n=50)
0.40*0.02 (n=18)
0.38*0.02 (n=32)
Dopamine (30 pmol/l)
0.15-0.46 (n=29)
0.27-10.01 (n= 12)
0.26 -t 0.02 (n= 17)
HMPG (800 pmol/l)
0.18-0.54 (n=30)
0.33*0.01 (n= 12)
0.32%0.02 (n= 18)
Phenol (30 fimol/l)
0.03-O. 19 (n=50)
0.07-10.01
0.07-0.01
(n=
(n=321
18)
AND
MEAN
341
Correlation of platelet PST activities with different substrates There was a high degree of correlation between platelet PST activities with tyramine and dopamine, tyramine and HMPG and dopamine and HMPG (Table II). However, there was no significant correlation between platelet PST activities with tyramine and phenol, dopamine and phenol or HMPG and phenol (Table II). Platelet PST activity range for different substrates The platelet PST activity range also varied for the different substrates used: for tyramine, dopamine and HMPG, there was an approximately 3-fold variation (Table III). However, when phenol was used as substrate, there was an approximately 6-fold variation of platelet activity in the same population (Table III). Effect of age and sex on platelet PST activity In this group of subjects, there was no significant difference in platelet PST activity between males and females for any of the substrates used (Table III). The age of the subjects varied from 20-82 years but there was no correlation of PST activity with age for tyramine as substrate (correlation coefficient = -0.06), dopamine (correlation coefficient = -0.22), HMPG (correlation coefficient = -0.07) or phenol (correlation coefficient = 0.07). Correlation between platelet PST and MAO activities There was no correlation between platelet PST activity with tyramine or phenol as substrate and MAO activity using tyramine as substrate in the 29 subjects for whom platelet monoamine oxidase activity was measured (correlation coefficient = 0.25 and - 0.04 respectively). Discussion PST assay The results described here show that great care must be taken when preparing samples for PST assay for clinical studies. In particular, enzyme thermolability and lability on storage in isotonic saline need to be noted. However, it does appear to be stable at - 20°C for several months when stored in 10 mmol/l phosphate buffer, pH 7.4, and it can also be frozen and thawed several times without loss of activity if kept in this way. Anderson and Weinshilboum [ 131 previously found that platelet PST is 50% inhibited by 0.1 mol/l NaCl and the slightly higher ionic strength we employed may well have contributed to our own findings. When platelet PST was prepared under optimal conditions, BSA and DTT had no significant effect on its activity; thus there was no advantage in including them in the incubation medium. Other workers [6], noted a 4-fold increase in PST activity when they added BSA and DTT at concentrations similar to those we investigated. However, the platelet suspension they used was substantially more dilute than ours, a factor which may contribute to greater enzyme instability. These investigations were all carried out using tyramine as substrate and so only apply to the “M” form of the enzyme, since this was the form in which we were primarily interested for clinical studies. However, it appears
342
that PST “P” is more Sandler, in preparation).
thermostable
than
PST “M”
(G. Rein,
V. Glover
and
M.
PST in depression This study has shown that low conjugated tyramine output in depression [l-3] cannot be accounted for by a generalised deficiency in PST itself. The mean and range of PST activity values, using tyramine as substrate, were similar in patients and controls. The patients investigated were of a diagnostic group previously found to exhibit the greatest conjugation deficit [2] and included one subject actually tested in this way on a previous occasion and shown to have low values. This patient had a PST activity value close to the mean of the staff control group (0.38 nmol product formed/mg protein/l0 min). The mean and range of PST values for depressed patients, patient controls and staff controls were also very similar when phenol and HMPG were used as substrates. With dopamine, the mean for the depressed patients was significantly less than the patient controls ( p c 0.05) although not significantly less than staff controls. It seems possible that this is a spurious finding: the group of depressed patients tested with dopamine was much smaller (n = 9) than that tested with tyramine (n = 18) and, as we have discussed, there is good evidence that both these substrates are sulphoconjugated by the same enzyme, PST “M”, as is HMPG. Several groups have found platelet MAO activity to be slightly but significantly raised in unipolar depression [ 141. In the small sample we investigated, MAO activity was somewhat higher in depressed patients and patient controls than staff controls but not significantly so. If decreased sulphoconjugation in depression were explicable in terms of a response to high MAO activity [l], one would have expected the increase to be substantially greater. Although these findings rule out a mutant PST “M” enzyme as the cause of the conjugation defect in depression, it does remain possible that intestinal activity, quantitatively one of the most important sources of the enzyme in the body [9], is for some reason reduced in these patients. Endogenous PST inhibitors have been demonstrated [13] and their concentration might vary in particular tissues under different circumstances. The assay conditions used here for measuring platelet activity would be unlikely to detect the presence of reversible inhibitors which would be greatly diluted in the assay mixture. However, other explanations are possible, including reduced tyramine uptake at sites of conjugation or a deficiency of the sulphate donor, PAPS. Independent control of PST “M” and “P” We thought it legitimate to combine all the data from patients and control subjects to determine whether PST “M” and “P” are controlled together or independently since they showed no significant difference in PST activity with any of the substrates used. The high degree of correlation between PST activities in platelets from different individuals using tyramine, dopamine and HMPG as substrates confirms that these three substances are sulphated by the same form of the enzyme, PST “M”. Anderson et al. [ 151 have obtained similar results using these substrates. On the contrary, there
343
is no correlation between phenol conjugation and that of the other substrates, providing strong new evidence that phenol is sulphoconjugated by a different form of the enzyme, PST “P”. It also demonstrates that the “P” enzyme is controlled independently from the “M” form. In addition, the two platelet PST enzymes are controlled separately from monoamine oxidase. Larger numbers of subjects will be needed to determine whether the range of activities of the platelet “P” enzyme is consistently greater than that of the “M” form. Here we have found a 3-fold variation among 30-50 subjects for the “M” enzyme and a 6-fold variation for the “P” enzyme. Anderson and Weinshilboum [6] found a 5-fold variation among 75 subjects, using HMPG as substrate and presumably measuring “M” enzyme activity. They similarly failed to find any significant difference in platelet PST “M” activity between males and females, using HMPG as substrate. We confirm this finding for the three “M” substrates we have used and were unable to identify any sex difference for “P” activity either. There was no change in activity of either enzyme with age. Acknowledgements
We would like to thank Miss Julia Littlewood for measuring platelet monoamine oxidase activity, Mrs. Anne Darby-Dowman for statistical advice, Mr. Glen Rein for advice and discussion and the nursing staff of the Brook General Hospital for their help in collecting the blood samples. Dr. Vivette Glover was supported by a grant from the Parkinson’s Disease Society. References 1 Sandier M, Bonham Carter S, Cuthbert MF, Pare CMB. Is there an increase in monoamine-oxidase activity in depressive illness? Lancet i 1975; 1045-1049. 2 Bonham Carter S, Sandier M, Goodwin BL. Sepping P, Bridges PK. Decreased urinary output of tyramine and its metabolites in depression. Br J. Psychiat 1978; 132: 125- 132. 3 Bonham Carter SM, Reveley MA, Sandier M, Dewhurst J, Little BC, Hayworth J. Priest RG. Decreased urinary output of conjugated tyramine is associated with lifetime vulnerability to depressive illness. Psychiat Res 1980; 3: 13-21. 4 Sandler M, Usdin E, eds. Phenolsulfotransferase in mental health research. Basingstoke: Macmillan, 1981. 5 Hart RF, Renskers KJ, Nelson EB, Roth JA. Localization and characterization of phenol sulphotransferase in human platelets. Life Sci 1979; 24: 125- 130. 6 Anderson RJ, Weinshilboum RM. Phenolsulphotransferasein human tissue: radiochemical enzymatic assay and biochemical properties. Clin Chim Acta 1980; 103: 79-90. 7 Rein G, Glover V, Sandler M. Sulphate conjugation of biologically active monoamines and their metabolites by human platelet phenolsulphotransferase. Clin Chim Acta 198 I ; I I I : 247-256. 8 Williams RT. Detoxication mechanisms. London: Chapman and Hall, 1959. 9 Rein G, Glover V, Sandier M. Phenolsulfotransferase in human tissues: evidence for multiple forms, In: Sandler M, Usdin E, eds. Phenolsulfotransferase in Mental Health Research. Basingstoke: Macmillan, 1981: 98-126. IO Foldes A, Meek JL. Rat brain phenolsulphotransferase-partial purification and some properties. Biochim Biophys Acta 1973; 327: 365-373. I I Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951; 193: 265-273.
344
12 Eisenthal R, Comish-Bowden A. The direct linear plot. A new graphical procedure for estimating enzyme kinetic parameters. Biochem J 1974; 139: 7 15- 720. 13 Anderson RJ, Weinshilboum RM. Phenolsulphotransferase: enzyme activity and endogenous inhibitors in the human erythrocyte. J Lab Clin Med 1979; 94: 158- 17 I, 14 Reveley MA, Clover V, Sandier M, Coppen A. Increased platelet monoamine oxidase activity in affective disorders. Psychopharmacology 198 1; 73: 257- 270. 15 Anderson RJ, Weinshilboum RM, Phillips SF, Broughton DD. Human platelet phenol sulpho\ransferase: assay procedure, substrate and tissue correlations. Clin Chim Acta 1981; 110: 157- 168.
Clinicu Chimicu Actu, 117 (1981) 333-344 Elsevier/North-Holland Biomedical Press
CCA
1978
Human platelet phenolsulphotransferase: separate control of the two forms and activity range in depressive illness Susan
M. Bonham
Carter
a, Vivette
P.K. Gillman
Glover
b and P.K.
a, M. Sandler
Bridges
a,*,
b
u Bernhurd Buwn Memoriul Reseurch Luhorutories und Institute of Obstetrics and Gynuecologv, Queen Charlotte’s Hospitul, Goldhuwk Roud, London Wci OXG (UK) und’ The Geoffrey Knight Psychosurgicul Unit, Brook Generul Hospitul, Shooters Hill Roud, London SE18 4L W (UK) (Received
June 16th. 198 I)
Summary Human phenolsulphotransferase exists in two forms, one specific for dopamine and tyramine, termed “M” and one for phenol, termed “I”‘. In this study we have shown that these two forms are under separate control by correlating their activities in different individuals using different substrates. There was a highly significant correlation between the activities with dopamine, p-tyramine and 4-hydroxy-3methoxyphenylglycol, but no significant correlation between the activities with any of these three substrates and that with phenol. Neither age nor sex had any effect on platelet phenolsulphotransferase “M” or “P” activities. Nor was there any significant correlation between platelet monoamine oxidase activity and phenolsulphotransferase “M” or “I”’ activities. Human platelet phenolsulphotransferase “M” was found to be unstable at temperatures above 35°C and it lost substantial activity when stored deep frozen in isotonic saline. However it was stable for up to four months when stored in isotonic sucrose or 10 mmol/l phosphate buffer (pH 7.4). Phenolsulphotransferase “M” and “I”’ activities were measured in platelets from depressed patients of a diagnostic type characterized by low output of tyramine-osulphate after oral tyramine loading but their enzyme activities were not different from those in two control groups.
Introduction Depressed patients tend to excrete less conjugated tyramine (tyramine-0-sulphate) after an oral load than normal subjects [ 1,2] and this defect is trait- rather than * To whom correspondence 0009-8981/81/0000-0000/$02.75
should be addressed. 0 1981 Elsevier/North-Holland
Biomedical
Press
334
state-dependent [2,3]. One possible cause is a decrease in activity of phenolsulphotransferase (PST) (EC 2.8.2.1) the enzyme which catalyzes the transfer of sulphate from 3’-phosphoadenosine-S-phosphosulphate (PAPS) to a phenolic acceptor [4]. PST has recently been identified in human platelets [5-71, thus opening the way to a wide variety of clinical studies. In this present work, we have measured PST activity in severely depressed patients, patient controls and staff controls. Because little work has so far been done with human PST, it was first necessary to establish the best conditions for platelet preparation and storage prior to enzyme assay. Sulphate conjugation catalyzed by PST is an important metabolic pathway, not only for phenolic monoamines and some of their metabolites but also for phenol and some phenolic drugs [8]. Rein et al. [9] have recently found evidence for the existence of two forms of human PST, one specific for phenol (named PST “P”) and one for tyramine and dopamine (PST “M”). They differ in tissue distribution and inhibitor sensitivity. In order to determine whether these two forms are controlled together or independently, we have determined the degree of correlation between platelet PST activities in individual subjects using tyramine, dopamine, 4 - hydroxy - 3 methoxyphenylglycol (HMPG) and phenol as substrates. Materials [35S]PAPS (1.9-4.2 Ci/mmol; New England Nuclear Corporation, Boston, MA, USA) was shipped in dry ice and stored in 50% ethanol at -20°C. Non-radioactive PAPS (PL Biochemicals, Inc., Milwaukee, WI, USA) was shipped and stored similarly. p-Tyramine HCl, dopamine HCl, 4-hydroxy-3-methoxyphenylglycol piperazine salt (HMPG), dithiothreitol (DTT) and bovine serum albumin (BSA) were purchased from Sigma Chemical Company, Poole, Dorset, UK. Phenol was obtained from BDH Chemicals Ltd., Poole, Dorset, UK and Instagel from Packard Instrument Co., Wembley, UK. Methods Sample
collection and platelet preparation
For stability studies: venous blood samples (10 ml) were collected from normal volunteers into plastic Universal containers with 0.5 ml 5% ethylenediaminetetraacetic acid (EDTA) and mixed gently. They were centrifuged at 327 X g for 6 min at room temperature; platelet-rich plasma was carefully removed by aspiration and samples from groups of three or four subjects were pooled before being divided equally into three and spun again at 2500 X g for 20 min at room temperature. The three platelet buttons so obtained were treated differently; one was resuspended in isotonic saline (0.15 mol/l; 1 ml/10 ml blood) and two in isotonic sucrose (0.3 mol/l; 1 ml/10 ml blood) before transfer to another Universal container, care being taken to leave any red blood cells behind. The suspensions were recentrifuged at 2500 X g for 20 min. The platelet button that had been washed with saline was
335
drained and resuspended in isotonic saline (1 ml/ 10 ml blood); one of the preparations washed with isotonic sucrose was resuspended in isotonic sucrose (1 ml/10 ml blood) and the other in phosphate buffer (10 mmol/l, pH 7.4; 1 ml/10 ml blood). These suspensions were then divided into aliquots as required before storage at - 20°C. For clinical studies: blood was obtained from 18 patients suffering from therapyresistant depressive illness, of the same diagnostic category in which we had already demonstrated a decreased tyramine-conjugating ability [2], from 18 non-psychiatric patient control subjects and 14 staff control subjects. Blood samples from 30 of the subjects (10 depressed patients, 10 patient controls and 10 staff controls) were collected on one day and processed together, and from the remaining 20 subjects (8 depressed patients, 8 patient controls and 4 staff controls) on another day, again being processed together. Venous blood (10 ml from each subject) was collected into plastic Universal containers with 0.5 ml 5% EDTA and mixed gently. It was centrifuged at 327 X g for 6 min, the platelet-rich plasma being carefully transferred to a new Universal container and the platelets spun down at 2500 X g for 20 min. Platelet buttons were drained, resuspended in isotonic sucrose (1 ml/10 blood) and transferred to new Universal containers, care being taken to leave any red blood cells behind (washing step). The platelets were spun down again at 2500 X g for 20 min. After the supernatant had been removed, the platelets were finally resuspended in phosphate buffer (10 mmol/l, pH 7.4; 1 ml/ 10 ml blood) and stored at - 20°C until assay. PST
assay
PST activity was measured by the radioenzymatic method of Foldes and Meek [lo], using radioactive [35S]PAPS as sulphate donor. For stability studies, the incubation mixture consisted of 100 ~1 potassium phosphate buffer (10 mmol/l, pH 7.4) 10 ~1 pooled platelet suspension and 20 ~1 tyramine solution (1 mmol/l, giving a final concentration of 133 pmol/l in the incubation mixture) or 20 ~1 water for the blanks. All samples were assayed in duplicate with a blank for each pooled platelet sample. The reaction was initiated by adding 20~1 of a 4.5 pmol/l solution of [35S]PAPS and non-radioactive PAPS in water (final concentration in incubation mixture 0.6 pmol/l) to successive tubes at 10 s intervals. After incubating for 10 min at 37°C the reaction was stopped at 10 s intervals by adding 200 ~1 barium acetate (0.1 mol/l) and transferring both from the water to ice. Unreacted PAPS was removed by two successive precipitations with barium hydroxide (200~1, 0.1 mol/l) and zinc sulphate (200 ~1, 0.1 mol/l), the second being carried out on the supernatant obtained from spinning down the first. After centrifugation, the final supernatant was removed to a counting vial insert; Instagel (2.5 ml) was added and mixed to form a gel and the radioactivity measured in a Packard Tricarb liquid scintillation counter. Results were expressed as nmol product (tyramine-O-sulphate) formed/mg protein/l0 min incubation. For the clinical study, the assay procedure was the same but it was also performed on the platelet samples using acceptor substrates other than tyramine in the following final concentrations: phenol at 30 pmol/l, dopamine at 30 pmol/l and
336
HMPG at 800 pmol/l; 20~1 platelet suspension was used in the assay with phenol because of lower specific activity. Blanks containing 204 of water instead of acceptor substrate were assayed for each platelet sample. This was essential as blank values varied from individual to individual, although they were no more than 20% of the total using tyramine as substrate. All assays were carried out in duplicate, with the operator blind to the clinical data. Monoamine oxidase (MAO) assay MAO activity towards tyramine was measured in the platelets obtained from the clinical study, using a standard radioenzymatic method with [‘4C]tyramine as substrate (final concentration 150 pmol/l). Protein determination Protein concentrations of platelet suspensions were measured by the method of Lowry et al [l l] with BSA as standard. Results
Linearity with time and tissue concentration Platelet PST activity towards tyramine increased linearly with time up to 10 min incubation at 37°C. A 10 min incubation period was therefore used in all assays. Activity also increased linearly with increasing tissue concentration up to a value of
-60
-50
-40
-30
-20
-10
IO
20
30
Fig. 1. Double reciprocal plots of human platelet PST activity versus tyramine concentration at constant concentrations of PAPS: 0 0.06 pmol/l; 0 0.12 pmol/l; n 0.24 gmol/l; A 0.6 pmol/l and H 1.2 amol/l. K, for tyramine obtained from this experiment= 15 pmol/l; a second experiment gave similar results. Lines were reconstructed and the K, calculated by the direct linear plot method [121.
337
60 pg protein in the incubation mixture. The actual amount of protein used in assays ranged from IO-60 pg. K,
determinations
Human platelet PST activity was measured using different concentrations of tyramine and PAPS. The results are shown in Figs. 1 and 2. Apparent K, values were calculated by the direct linear plot method [ 121.The apparent K, for tyramine was 17 pmol/l (mean of two experiments), a value somewhat lower than that obtained by others, using Lineweaver-Burk plots [5,7]. The apparent K, for PAPS was 0.14 pmol/l (mean of two experiments), very similar to that obtained by Anderson and Weinshilboum [6] who also used the direct linear plot. Effect of BSA, of DTT and of washing the platelet preparation
Addition of BSA (final concentration 0.042%) and DTT (final concentration 8 mmol/l) to the incubation mixture had no significant effect on PST activity towards tyramine of pooled human platelets suspended in isotonic saline, isotonic sucrose or 10 mmol/l phosphate buffer. However, as will be seen in the section on stability with storage, DTT did sometimes partially restore PST activity which had been lost when the suspension was stored deep frozen in isotonic saline. Washing the platelet button, as described in the “Methods” section, was found to be essential to remove plasma proteins.
&
(gmol/l-‘)
Fig. 2. Double reciprocal plots of human platelet PST activity versus PAPS concentration at constant concentrations of tyramine: 0 33 pmol/l; 0 67 pmol/l; n 167 pmol/l. K, for PAPS obtained in this experiment=O. I5 pmol/l; a second experiment gave similar results. Lines were reconstructed and the K, calculated by the direct linear plot method [ 121.
Fig. 3. Thermostability of human platelet PST using tyramine as substrate. Platelets saline, n ; isotonic sucrose, A ; and phosphate buffer (10 mmol/l, pH 7.4), 0.
suspended
in isotonic
Freezing and thawing Different aliquots of the same pooled platelet sample, suspended in isotonic saline, isotonic sucrose or 10 mmol/l phosphate buffer, were frozen and thawed up to five times before assay for PST activity towards tyramine. Samples suspended in saline lost up to 50% of activity after such treatment and it was not restored by adding DTT (final concentration 8 mmol/l) to the incubation mixture. Platelets suspended in sucrose lost up to 25% activity after this repeated manoeuvre and, again, DTT did not restore lost activity. Platelets stored in 10 mmol/l phosphate buffer, however, lost no appreciable PST activity after successive freezing and thawing. Thermostability Aliquots of pooled platelet samples suspended in isotonic saline, isotonic sucrose and 10 mmol/l phosphate buffer were exposed to different temperatures in a water bath for 20 min prior to PST assay using tyramine as substrate. The results are shown in Fig. 3. The enzyme loses most of its activity at 40°C and a substantial amount even at 35°C in saline, sucrose or buffer; again, the addition of DTT to the incubation mixture failed to restore lost activity. Thus, it is obvious that great care must be taken when preparing platelets for PST studies, not allowing them to overheat, for instance, during centrifugation.
339 PST activity I-
60
80
100
(days)
Fig. 4. Stability of human platelet PST activity with storage time, using tyramine as substrate. Platelets suspended in isotonic saline, m; isotonic sucrose, A; and phosphate buffer (IO mmol/l, pH 7.4) l .
Stability with storage time Successive aliquots of pooled platelet samples stored deep frozen (-20*C) in isotonic saline, isotonic sucrose or 10 mmol/l phosphate buffer were assayed for PST activity using tyramine as substrate over a period of 116 days. No aliquot was thawed more than once prior to assay. The results are shown in ‘Fig. 4. It will be seen that, whereas samples stored in sucrose and phosphate buffer were stable over that time period, those stored in saline lost more than 85% of their activity after 56 days. The addition of DTT (final concentration 8 mmol/l) to the incubation mixture had no effect on PST activity of platelets stored in sucrose or buffer and only partially restored the lost activity of platelets stored in saline. PST activity in depressive illness Once the stability of human platelet PST in the three different platelet suspension media had been established, blood samples were collected from severely depressed patients and controls and platelet suspensions prepared and stored as described in the “Methods” section. PST activity was assayed in all 50 samples using tyramine TABLE
I
HUMAN = SEM)
PLATELET
PST ACTIVITIES
Substrate
(nmol product
formed/mg
protein/
IO min incubation:
Staff controls
Patient controls
Depressed patients
Tyramine
(133tr,mol/l)
0.39*0.03 (n=14)
0.39~0.02 (n=l8)
0.38*0.02 (n=ltl)
Dopamine
(30 pmol/l)
0.28t0.03 (?I= IO)
0.28r0.01 (n= 10)
0.23 20.02 (n=9f
0.33 eo.03 (n= IO)
0.32*0.02 (n=lO)
0.32 ItO. (I?= IO)
0.07 k 0.01 (n= 14)
0.07~0.01 (n=l@
0.07 2 0.01 (n== 18)
HMPG
(800 amol/l)
Phenol (30 amol/I)
* pCO.05
compared
to patient
controls
(Student
r-test).
*
mean
340
TABLE
II
CORRELATION INDIVIDUALS
COEFFICIENTS FOR HUMAN PLATELET WITH DIFFERENT SUBSTRATES
PST ACTIVITIES
IN DIFFERENT
Substrates
Tyramine
Dopamine
HMPG
Phenol
Tyramine
_
0.78 * (n=29)
0.88 *
0.27
(n=30)
(n=SO)
0.78 *
0.82 *
0.19
(n=29)
(n=29)
(n=29)
_
0.24 (n=30)
Dopamine
HMPG
0.82 * (n=29)
0.88 * (n=30)
* p
III
HUMAN VALUES
PLATELET PST: ACTIVITY RANGES FOR MALES AND FEMALES
Substrate (final concentration)
PST activity (nmol product Activity
range
formed/mg
FOR
DIFFERENT
protein/
SUBSTRATES
IO min)
Mean -ISEM males
females
Tyramine (133 pmol/l)
0.20-0.63 (n=50)
0.40*0.02 (n=18)
0.38*0.02 (n=32)
Dopamine (30 pmol/l)
0.15-0.46 (n=29)
0.27-10.01 (n= 12)
0.26 -t 0.02 (n= 17)
HMPG (800 pmol/l)
0.18-0.54 (n=30)
0.33*0.01 (n= 12)
0.32%0.02 (n= 18)
Phenol (30 fimol/l)
0.03-O. 19 (n=50)
0.07-10.01
0.07-0.01
(n=
(n=321
18)
AND
MEAN
341
Correlation of platelet PST activities with different substrates There was a high degree of correlation between platelet PST activities with tyramine and dopamine, tyramine and HMPG and dopamine and HMPG (Table II). However, there was no significant correlation between platelet PST activities with tyramine and phenol, dopamine and phenol or HMPG and phenol (Table II). Platelet PST activity range for different substrates The platelet PST activity range also varied for the different substrates used: for tyramine, dopamine and HMPG, there was an approximately 3-fold variation (Table III). However, when phenol was used as substrate, there was an approximately 6-fold variation of platelet activity in the same population (Table III). Effect of age and sex on platelet PST activity In this group of subjects, there was no significant difference in platelet PST activity between males and females for any of the substrates used (Table III). The age of the subjects varied from 20-82 years but there was no correlation of PST activity with age for tyramine as substrate (correlation coefficient = -0.06), dopamine (correlation coefficient = -0.22), HMPG (correlation coefficient = -0.07) or phenol (correlation coefficient = 0.07). Correlation between platelet PST and MAO activities There was no correlation between platelet PST activity with tyramine or phenol as substrate and MAO activity using tyramine as substrate in the 29 subjects for whom platelet monoamine oxidase activity was measured (correlation coefficient = 0.25 and - 0.04 respectively). Discussion PST assay The results described here show that great care must be taken when preparing samples for PST assay for clinical studies. In particular, enzyme thermolability and lability on storage in isotonic saline need to be noted. However, it does appear to be stable at - 20°C for several months when stored in 10 mmol/l phosphate buffer, pH 7.4, and it can also be frozen and thawed several times without loss of activity if kept in this way. Anderson and Weinshilboum [ 131 previously found that platelet PST is 50% inhibited by 0.1 mol/l NaCl and the slightly higher ionic strength we employed may well have contributed to our own findings. When platelet PST was prepared under optimal conditions, BSA and DTT had no significant effect on its activity; thus there was no advantage in including them in the incubation medium. Other workers [6], noted a 4-fold increase in PST activity when they added BSA and DTT at concentrations similar to those we investigated. However, the platelet suspension they used was substantially more dilute than ours, a factor which may contribute to greater enzyme instability. These investigations were all carried out using tyramine as substrate and so only apply to the “M” form of the enzyme, since this was the form in which we were primarily interested for clinical studies. However, it appears
342
that PST “P” is more Sandler, in preparation).
thermostable
than
PST “M”
(G. Rein,
V. Glover
and
M.
PST in depression This study has shown that low conjugated tyramine output in depression [l-3] cannot be accounted for by a generalised deficiency in PST itself. The mean and range of PST activity values, using tyramine as substrate, were similar in patients and controls. The patients investigated were of a diagnostic group previously found to exhibit the greatest conjugation deficit [2] and included one subject actually tested in this way on a previous occasion and shown to have low values. This patient had a PST activity value close to the mean of the staff control group (0.38 nmol product formed/mg protein/l0 min). The mean and range of PST values for depressed patients, patient controls and staff controls were also very similar when phenol and HMPG were used as substrates. With dopamine, the mean for the depressed patients was significantly less than the patient controls ( p c 0.05) although not significantly less than staff controls. It seems possible that this is a spurious finding: the group of depressed patients tested with dopamine was much smaller (n = 9) than that tested with tyramine (n = 18) and, as we have discussed, there is good evidence that both these substrates are sulphoconjugated by the same enzyme, PST “M”, as is HMPG. Several groups have found platelet MAO activity to be slightly but significantly raised in unipolar depression [ 141. In the small sample we investigated, MAO activity was somewhat higher in depressed patients and patient controls than staff controls but not significantly so. If decreased sulphoconjugation in depression were explicable in terms of a response to high MAO activity [l], one would have expected the increase to be substantially greater. Although these findings rule out a mutant PST “M” enzyme as the cause of the conjugation defect in depression, it does remain possible that intestinal activity, quantitatively one of the most important sources of the enzyme in the body [9], is for some reason reduced in these patients. Endogenous PST inhibitors have been demonstrated [13] and their concentration might vary in particular tissues under different circumstances. The assay conditions used here for measuring platelet activity would be unlikely to detect the presence of reversible inhibitors which would be greatly diluted in the assay mixture. However, other explanations are possible, including reduced tyramine uptake at sites of conjugation or a deficiency of the sulphate donor, PAPS. Independent control of PST “M” and “P” We thought it legitimate to combine all the data from patients and control subjects to determine whether PST “M” and “P” are controlled together or independently since they showed no significant difference in PST activity with any of the substrates used. The high degree of correlation between PST activities in platelets from different individuals using tyramine, dopamine and HMPG as substrates confirms that these three substances are sulphated by the same form of the enzyme, PST “M”. Anderson et al. [ 151 have obtained similar results using these substrates. On the contrary, there
343
is no correlation between phenol conjugation and that of the other substrates, providing strong new evidence that phenol is sulphoconjugated by a different form of the enzyme, PST “P”. It also demonstrates that the “P” enzyme is controlled independently from the “M” form. In addition, the two platelet PST enzymes are controlled separately from monoamine oxidase. Larger numbers of subjects will be needed to determine whether the range of activities of the platelet “P” enzyme is consistently greater than that of the “M” form. Here we have found a 3-fold variation among 30-50 subjects for the “M” enzyme and a 6-fold variation for the “P” enzyme. Anderson and Weinshilboum [6] found a 5-fold variation among 75 subjects, using HMPG as substrate and presumably measuring “M” enzyme activity. They similarly failed to find any significant difference in platelet PST “M” activity between males and females, using HMPG as substrate. We confirm this finding for the three “M” substrates we have used and were unable to identify any sex difference for “P” activity either. There was no change in activity of either enzyme with age. Acknowledgements
We would like to thank Miss Julia Littlewood for measuring platelet monoamine oxidase activity, Mrs. Anne Darby-Dowman for statistical advice, Mr. Glen Rein for advice and discussion and the nursing staff of the Brook General Hospital for their help in collecting the blood samples. Dr. Vivette Glover was supported by a grant from the Parkinson’s Disease Society. References 1 Sandier M, Bonham Carter S, Cuthbert MF, Pare CMB. Is there an increase in monoamine-oxidase activity in depressive illness? Lancet i 1975; 1045-1049. 2 Bonham Carter S, Sandier M, Goodwin BL. Sepping P, Bridges PK. Decreased urinary output of tyramine and its metabolites in depression. Br J. Psychiat 1978; 132: 125- 132. 3 Bonham Carter SM, Reveley MA, Sandier M, Dewhurst J, Little BC, Hayworth J. Priest RG. Decreased urinary output of conjugated tyramine is associated with lifetime vulnerability to depressive illness. Psychiat Res 1980; 3: 13-21. 4 Sandler M, Usdin E, eds. Phenolsulfotransferase in mental health research. Basingstoke: Macmillan, 1981. 5 Hart RF, Renskers KJ, Nelson EB, Roth JA. Localization and characterization of phenol sulphotransferase in human platelets. Life Sci 1979; 24: 125- 130. 6 Anderson RJ, Weinshilboum RM. Phenolsulphotransferasein human tissue: radiochemical enzymatic assay and biochemical properties. Clin Chim Acta 1980; 103: 79-90. 7 Rein G, Glover V, Sandler M. Sulphate conjugation of biologically active monoamines and their metabolites by human platelet phenolsulphotransferase. Clin Chim Acta 198 I ; I I I : 247-256. 8 Williams RT. Detoxication mechanisms. London: Chapman and Hall, 1959. 9 Rein G, Glover V, Sandier M. Phenolsulfotransferase in human tissues: evidence for multiple forms, In: Sandler M, Usdin E, eds. Phenolsulfotransferase in Mental Health Research. Basingstoke: Macmillan, 1981: 98-126. IO Foldes A, Meek JL. Rat brain phenolsulphotransferase-partial purification and some properties. Biochim Biophys Acta 1973; 327: 365-373. I I Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951; 193: 265-273.
344
12 Eisenthal R, Comish-Bowden A. The direct linear plot. A new graphical procedure for estimating enzyme kinetic parameters. Biochem J 1974; 139: 7 15- 720. 13 Anderson RJ, Weinshilboum RM. Phenolsulphotransferase: enzyme activity and endogenous inhibitors in the human erythrocyte. J Lab Clin Med 1979; 94: 158- 17 I, 14 Reveley MA, Clover V, Sandier M, Coppen A. Increased platelet monoamine oxidase activity in affective disorders. Psychopharmacology 198 1; 73: 257- 270. 15 Anderson RJ, Weinshilboum RM, Phillips SF, Broughton DD. Human platelet phenol sulpho\ransferase: assay procedure, substrate and tissue correlations. Clin Chim Acta 1981; 110: 157- 168.